National crime trend aggregates built from mostly large‑city reporting can show directionally useful signals but conceal suburban, rural, and intra‑metro dynamics that are necessary to adjudicate causal explanations (policing tactics, economic change, demographics). Without a more representative, geographically disaggregated and timely dataset, policymakers will be flying blind when deciding which interventions to scale.
— If true, fixing crime data coverage is a prerequisite for evidence‑based justice policy because the national decline could rest on localized drivers with very different policy remedies.
2026.01.05
100% relevant
Yglesias cites the Real Time Crime Index (377 agencies, big‑city heavy) and explicitly notes Georgia’s limited participating agencies (Atlanta, Athens, Dunwoody) as a concrete example of the sampling problem.
← Back to All Ideas